Research
My research consists of the use of human-centered artificial intelligence (AI) to develop novel digital diagnostics, screening tools, therapeutics, and interventions for a wide range of health conditions that I or my trainees care about. Our projects are focused on consumer digital health informatics, or the use of data from consumer devices such as smartphones, smartwatches, other wearables (think Oura ring for example), and consumer health devices (think blood pressure or glucose monitors for example) to derive actionable health insights.
​
More details can be found on my UCSF profile page (COMING SOON) and my research lab website: the UCSF TECH (Technologies Empowering Consumer Health) Lab (COMING SOON).
Human-In-The-Loop AI for Diagnostics
Through our NIH-funded DP2 grant, my lab is studying how humans and AI can work together to achieve diagnostic outcomes that would not be possible or feasible using just AI or humans, respectively. We are applying these methods in particular to psychiatry and behavioral sciences. We collaborate closely with the Wall Lab at Stanford and University of Hawaii on this project.
Personalized AI for Interventions and Therapeutics
Through our NSF-funded Smart and Connected Health grant, we are developing AI models that are optimized to make repeat predictions about an individual using data from consumer devices such as wearables and smartphones. We collaborate closely with researchers at Harvard, Ohio State, Kaiser Permanente, and University of Hawaii on this project.
.jpg)
Human-Centered AI
We are developing methods for fairness, privacy, and explainability as they relate to healthcare AI using a variety of data modalities, especially consumer digital devices but also including multimodal integrations with electronic health records and radiology imaging. We are working on building collaborations internally at UCSF and UC Berkeley to accomplish this line of work.